Lightnews — Scholar-powered news

Reposted by Susan Vanderplas

Susan Vanderplas @srvanderplas.bsky.social · 5d

nebraskapublicmedia.org/en/news/news...

#UNL is planning to eliminate the #statistics department using bad #statisticalAnalysis (srvanderplas.github.io/2025-stat-ap... -- it includes Cauchy variables, bad interpretation, and failure to ID a population).

Public Comment: apc.unl.edu/fall-2025-bu...

As $27.5 million in cuts loom over UNL, students hope to save programs

Students in programs up for elimination shared their concerns with how the loss of their programs would hurt the state. The potential cuts are in response to the state legislature allocating less fund...

nebraskapublicmedia.org

1

Reposted by Susan Vanderplas

Bruce Stallsmith @laimosemion.bsky.social · 3d

This may be the best response to much of the junk science being pushed by the trumpies in the US. 🧪🌎🐟⚒️ #TeamFish

7 31 110

Susan Vanderplas @srvanderplas.bsky.social · 5d

nebraskapublicmedia.org/en/news/news...

#UNL is planning to eliminate the #statistics department using bad #statisticalAnalysis (srvanderplas.github.io/2025-stat-ap... -- it includes Cauchy variables, bad interpretation, and failure to ID a population).

Public Comment: apc.unl.edu/fall-2025-bu...

As $27.5 million in cuts loom over UNL, students hope to save programs

Students in programs up for elimination shared their concerns with how the loss of their programs would hurt the state. The potential cuts are in response to the state legislature allocating less fund...

nebraskapublicmedia.org

1

Reposted by Susan Vanderplas

Dr Ellie Murray, ScD @epiellie.bsky.social · 6d

If you like data, birds, and making cool stuff, Jer is your guy! 👇🏼

Jer Thorp @jerthorp.bsky.social · 6d

I've opened up some membership tiers for Binoculars to Binomials.

If you'd like to support this work, get a monthly newsletter full of 🪶 + 📊 + 🎨 magic, and get in on new courses, trips and retreats from the ground floor, you can sign up:

binoculars-to-binomials.disco.co/memberships

Binoculars
to Binomials

A hybrid space for mindful learning about birds, data and creative possibility.

1 2 9

Reposted by Susan Vanderplas

Liz Neeley @lizneeley.bsky.social · 6d

Some problems have simple solutions. This new “Compact for Academic Excellence in Higher Education” is one. The answer is simply saying no.

Simple does not mean easy. But Brown, Dartmouth, MIT, Vanderbilt, Univ of AZ, UPenn, USC, UT Austin & UVA must understand what they’ll lose if they sign on.

2 190 490

Reposted by Susan Vanderplas

Emi Tanaka @emitanaka.org · 6d

Adrian is brave and courageous. He stood by his moral consciousness. Manipulating statistics for personal or political gain is a sure way of losing trust. Honestly, listen more to the statisticians.

1 5

Reposted by Susan Vanderplas

NE Ohio Regional Sewer District @neorsd.org · 7d

I’m a sewage system. I also have a great personality.

Elizabeth Jacobs, PhD
@elizabethjacobs.bsky.social
I'm an epidemiologist. Sewage systems are sexy.

Dosco Jones @doscojones.bsky.social
I'm an engineer. Sewage systems are sexy.

150 1.1K 8.6K

Reposted by Susan Vanderplas

Noam Ross @noamross.net · 7d

The Damage Done: @marianneguenot.bsky.social tells the story of clinical trials derailed by the administration's reckless science policy, across kidney disease, HIV, Alzheimer's disease, and cancer. Years of work and money thrown away, at times just short of a cure.

www.nature.com/articles/s41...

The damage done - Nature Medicine

When clinical trials are suddenly halted by US funding cuts, there are repercussions for investigators and patients, and they do not stop at borders.

www.nature.com

5 120 210

Susan Vanderplas @srvanderplas.bsky.social · 9d

Also, leave a comment in support of units you think it's crazy to cut here: apc.unl.edu/fall-2025-bu...

Feel free to suggest cutting the chancellor and/or president instead of departments like EDAD and Statistics (&others) that support NE's residents and economy!!

Fall 2025 Budget Reduction Feedback Form | Academic Planning Committee | Nebraska

apc.unl.edu

1

Reposted by Susan Vanderplas

Melodyvacc @melodyvacc.bsky.social · 9d

Save our university!

thebugeater.com/uncategorize...

You Can’t Do that With Statistics - The Bugeater

In my last post, I mentioned how incomplete Academic Analytics (AA) is at capturing every faculty member’s productivity. In this post, I’m going to be discussing the problems with how the AA data is b...

thebugeater.com

1 1 3

Susan Vanderplas @srvanderplas.bsky.social · 10d

It can still delete user files, though! Or grab data from your system and phone home.

1

Susan Vanderplas @srvanderplas.bsky.social · 10d

It's a bit terrifying to give AI the ability to run commands on your system!

1 6

Reposted by Susan Vanderplas

John Shirley @johnshirley2024.bsky.social · 10d

Does painting cows with stripes prevent fly bites? Researchers who studied this win Ig Nobel prize

A researcher who studied fingernail growth and a group that explored whether painting zebralike stripes on cows reduced fly bites are among the winners of this year's Ig Nobels.

apnews.com

16 46 270

Reposted by Susan Vanderplas

Grumpy Old Health Stats Dude @healthstatsdude.bsky.social · 12d

goodnight, “tweeps” (“bleeps”? “bluseeps”? “skeeps”?) #statistics #AI

1 4 19

Reposted by Susan Vanderplas

Claus Wilke @clauswilke.com · 12d

I asked my SRO. He didn't know, other than it's not happening at the level of the panel. Instead, some NIH staff make that determination. They probably run all applications through an AI detector, and if it flags an application as AI developed, too bad, the application is out.

Laura Williams @laurawilliams.bsky.social · 12d

New NIH policy on AI:

“Applications that are either substantially developed by AI or containing sections substantially developed by AI…will not be considered by NIH.”

No explanation of how NIH will identify AI-generated proposal text.

Dr. Michael D. L. Johnson @copperbae.bsky.social · 12d

Effective today grants.nih.gov/news-events/....

2 3 1

Susan Vanderplas @srvanderplas.bsky.social · 13d

There shouldn't be billionaires. That said, if they fund enough science to no longer be a billionaire that's ok I guess.

Reposted by Susan Vanderplas

Maëlle Salmon @masalmon.eu · 13d

"A Gentle Introduction to Open Science"

I love this post by Steffi LaZerte on @ropensci.org. Gentle indeed, and pragmatic.

ropensci.org/blog/2025/09...

A Gentle Introduction to Open Science

Steffi reviews a talk on open science. Don’t forget that it takes courage; be kind to yourself!

ropensci.org

1 13 26

Susan Vanderplas @srvanderplas.bsky.social · 13d

Recently proposed pitching an academic reality tv series about budget cuts as a fundraising opportunity. People don't get just how much drama there is in the academy.

1

Reposted by Susan Vanderplas

Aaron Sojourner @aaronsojourner.org · 13d

#priorities
www.propublica.org/article/geor...

Georgia’s Medicaid Work Requirement Program Spent Twice as Much on Administrative Costs as on Health Care, GAO Says
Republican lawmakers cite Georgia’s Pathways to Coverage as a national model for federal Medicaid work requirements that are set to take effect in 2027. A new report shows the program has spent at least $54 million on administrative costs alone.

1 35 84

Susan Vanderplas @srvanderplas.bsky.social · 13d

Also a bias in tendency to label things as autism because services are available for the diagnosis. There are a lot of factors here and they all muddy the waters.

1 1

Reposted by Susan Vanderplas

WOMBAT 2025 @wombat.numbat.space · Aug 28

If you're not 100% confident you can make it to her tutorial on visualising uncertainty, no doubt you won't want to miss Harriet Mason's talk on the ggdibbler package.

More info and registration link here: wombat2025.numbat.space/program/work...

#DataViz #RStats #Python #DataScience

WOMBAT 2025 - Invited talk

Harriet Mason - Visualising Uncertainty with ggdibbler

29-30 September 2025, Naarm/Melbourne
womabt2025.numbat.space

7 12

Susan Vanderplas @srvanderplas.bsky.social · 13d

This is me, as a statistician doing #visualization research and secretly not-quite-hating models.

2

Reposted by Susan Vanderplas

Dr Juulia Suvilehto @juulia.bsky.social · 14d

My mentor once said: strongest results are the ones where you don't even need stats to figure out if there was an effect, you just plot the raw data and look at it with your two human eyes. If it's obvious then, it's a ding-dang strong effect.

1 1 4

Reposted by Susan Vanderplas

Aaron Schiff @schiff.nz · 14d

Also … the better your data, the less fancy your analysis needs to be. You can do good work by spending a lot of effort to collect really good data.

Crystal Lewis @cghlewis.bsky.social · 14d

📣 PSA I've been wanting to say for a while.

Not everyone is an expert programmer or coder, and that's okay. People can still do good, reproducible research without using the fanciest methods or the coolest tools.

Low tech tools ≠ You can't do good work

3 2 14

Susan Vanderplas @srvanderplas.bsky.social · 14d

I'm currently fighting #AcademicAnalytics metrics being used to close my #stats program because administrators don't understand variance or context. Not quite the same as the college rankings you're talking about, but metrics rot is absolutely a concern all the way through the modern academy. 😭

4